Diversity of fungi in sediments and water sampled from the hot springs of Lake Magadi and Little Magadi in Kenya

1 Institute for Biotechnology Research, Jomo Kenyatta University of Agriculture and Technology, P. O. Box 62000 00200, Nairobi, Kenya. 2 Embu University College, P. O. Box 6 60100, Embu, Kenya. 3 International Centre of Insect Physiology and Ecology (ICIPE), P. O. Box 30772 00100, Nairobi, Kenya. 4 Pure and Applied Sciences, Technical University of Mombasa, P. O. Box 90420 80100, GPO, Mombasa, Kenya. 5 Taita Taveta University College, School of Agriculture, Earth and Environmental Sciences, P. O. Box 635-80300 Voi, Kenya.


INTRODUCTION
Fungi have colonized diverse habitats such as tropical regions (Hawksworth, 1991), extreme environments such as deserts, areas with high salt concentrations (Vaupotic et al., 2008), ionizing radiation (Dadachova et al., 2007), deep sea sediments (Raghukumar and Raghukumar, 1998) and ocean hydrothermal areas (Le Calvez et al., 2009).Most fungi grow in terrestrial environments, though several species live partly or solely in aquatic habitats, such as the chytrid fungus Batrachochytrium dendrobatidis, a parasite that has been responsible for a worldwide decline in amphibian populations (Brem and Lips, 2008).In most ecosystems, fungi are the major decomposers, playing an essential role in nutrient cycling as saprotrophs and symbionts that degrade organic matter into inorganic molecules (Barea et al., 2005;Lindahl et al., 2007;Gadd, 2007).While there are wellknown examples of bacteria that can grow in a variety of natural environments including hot springs and geysers where temperatures can reach 100°C, eukaryotes are much more sensitive because, above 65°C, their membranes become irreparably damaged (Magan and Aldred, 2007).However, mesophilic thermo-tolerant fungi exist.For example, some Deuteromycetes isolated from thermal springs have maximum growth temperature of 61.5°C (Magan, 2006).
The presence of fungi in extreme alkaline saline environments has been recognized by culture-dependent methods, with the majority showing similarity to terrestrial species (Mueller and Schmit, 2006;Salano, 2011;Ndwigah et al., 2015).Culture-independent methods have revealed highly novel fungal phylotypes such as Chytridiomycota and unknown ancient fungal groups (Yuriko and Takahiko, 2012).
pH tolerance in fungi has been attributed to efficient control of proton movement into and out of the cells, and is able to meet necessary energy requirements (Magan, 2006).The exact diversity and function of fungi in extreme environments is still poorly understood.The aim of this study was to explore the fungal diversity within the hot springs of Lake Magadi and Little Magadi in Kenya using metagenomic analysis.

Study site
Lake Magadi is a hyper saline lake that lies in a naturally formed closed lake basin within the Southern part of the Kenyan Rift Valley.It is approximately 2 °S and 36 °E of the Equator at an elevation of about 600 m above sea level (Behr and Röhricht, 2000).The solutes are supplied mainly by a series of alkaline springs with temperatures as high as 86°C, located around the perimeter of the lake.Samples analyzed in this study were collected from 3 hot springs: one hot spring within the main Lake Magadi (02° 00′ 3.7″S 36° 14′ 32″ E at an elevation of 603 m, a temperature of 45.1°C and pH 9.8), and two hot springs within Little Magadi "Nasikie eng'ida": Hot spring 1 -01° 43′ 28″ S 36° 16′ 21″ E, at an elevation of 611 m, a temperature of 83.6°C and pH 9.4 ); and Hot spring 2 -01° 43′ 56″ S 36° 17′ 11″ E, at an elevation of 616 m, temperature of 81°C and pH of 9.2 (Table 1).

Measurements of physicochemical parameters
Geographical position of each site in terms of latitude, longitude and elevation was taken using Global Positioning System (GARMIN eTrex 20).The pH for each sampling point was measured with a portable pH-meter (Oakton pH 110, Eutech Instruments Pty.Ltd) and confirmed with indicator strips (Merck, range 5-10), Temperature, Electrical Conductivity (EC), Total Dissolved Solids (TDS) and dissolved oxygen (DO) were measured on site using Electrical Chemical Analyzer (Jenway -3405) during sampling.In situ temperature was recorded once for each study site and assigned to all the sample types for that site.

Sample collection
All samples were collected randomly in triplicates from each hot spring.Water samples were collected using sterile 500 ml plastic containers that had been cleaned with 20% sodium hypochlorite and UV-sterilized for one hour.Wet sediments were collected by scooping with sterilized hand shovel into sterile 50 ml falcon tubes.All samples were transported in dry ice to the laboratory at Jomo Kenyatta University of Agriculture and Technology.Water for DNA extraction (500 ml) was trapped on 0.22 μM filter papers (Whatman) and stored at -80°C.Pellets for DNA extraction were obtained from water samples by re-suspending the filter papers in phosphate buffer solution (pH 7.5), and centrifuging 5 ml of the suspension at 13000 rpm for 10 min.

DNA extraction
Total community DNA was extracted in triplicates; pellets from water samples and 0.2 g of sediment samples as described by (Sambrook et al., 1989).The DNA extracted from triplicate samples was pooled during the precipitation stage, washed, air dried and stored at -20°C.

Amplicon library preparation and sequencing
PCR amplification of ITS region was done using ITS1 (TCCGTAGGTGAACCTGCGG) and TS4 (TCCTCCGCTTATTGATATGC) primers with barcode according to (White et al., 1990).Amplification proceeded in a 30 cycle PCR using the HotStarTaq Plus Master Mix Kit (Qiagen, USA) with initial heating at 94°C for 3 min, followed by 28 cycles of denaturation at 94°C for 30 s, annealing at 53°C for 40 s and extension at 72°C for 1 min, after which a final elongation step at 72°C for 5 min was performed.Polymerase chain reaction (PCR) products were visualized on 2% agarose gel to determine the success of amplification and the relative intensity of bands.Multiple samples were pooled together in equal proportions based on their DNA concentrations.Pooled samples were purified using calibrated Ampure XP beads (Agencourt Bioscience Corporation, MA, USA).The pooled and purified PCR product was used to prepare DNA library by following Illumina sequencing protocol (Yu and Zhang, 2012).Sequencing was performed at Molecular Research DNA (www.mrdnalab.com,Shallowater, TX, USA) on a MiSeq platform following the manufacturer's guidelines.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License

Sequence analysis, taxonomic classification and data Submission
Sequences obtained from the Illumina sequencing platform were depleted of barcodes and primers using a proprietary pipeline (www.mrdnalab.com,MR DNA, Shallowater, TX) developed at the service provider's laboratory.Low quality sequences were identified by denoising and filtered out of the dataset (Reeder and Knight, 2010).Sequences which were < 200 base pairs after phred20based quality trimming, sequences with ambiguous base calls, and those with homopolymer runs exceeding 6bp were removed.
Sequences were analyzed by a script optimized for high-throughput data to identify potential chimeras in the sequence files, and all definite chimeras were depleted as described previously (Gontcharova et al., 2010).De novo OTU clustering was done with standard UCLUST method using the default settings as implemented in QIIME pipeline Version 1.8.0 at 97% similarity level (Caporaso et al., 2010a).Taxonomy was assigned to each OTU using BLASTn against SILVA SSU Reference 119 database at default e-value threshold of 0.001 in QIIME (Quast et al., 2013).

Statistical analysis
Diversity indices (Shannon, Simpson and Evenness) for each sample were calculated using vegan package version 1.16-32 in R software version 3.1.3(R Development Core Team, 2012).Community and Environmental distances were compared using Analysis of similarity (ANOSIM) test, based upon Bray-Curtis distance measurements with 999 permutations.Significance was determined at 95% confidence interval (p=0.05).Calculation of Bray-Curtis dissimilarities between datasets and hierarchical clustering were carried out using the R programming language (R Development Core Team, 2012) and the Vegan package (Oksanen et al., 2012).To support OTU-based analysis, taxonomic groups were derived from the number of reads assigned to each taxon at all ranks from domain to genus using the taxa_summary.txtoutput from QIIME pipeline Version 1.8.0.Obtained sequences were submitted to NCBI Sequence Read Archive with SRP# Study accessions: SRP061806.

RESULTS
Wet sediment and water samples were randomly collected at three different locations in hot springs of Lake Magadi and Little Magadi.The hot springs temperatures ranged from 45.1 to 83.6°C while pH ranged from 9.2 to 9.8.The TDS was beyond measurement using the Electrical Chemical Analyzer; hence the readings appeared as one on the sampling equipment.The metadata collected before sampling is summarized in Table 1.Temperature measurement showed a gradient from hot spring in the main Lake Magadi, with the springs at Little Magadi measuring between 81 and 83.6°C.Cation analysis of the water samples showed that the levels of calcium range between 0.33-0.62ppm, iron (<0.01 -0.014 ppm) and magnesium (<0.02 -0.026 ppm).Sodium levels were very high (11,300, 17,300 and 17,700 ppm) and potassium levels were 225, 458 and 287 ppm.Anion analysis showed that phosphorus range between 2.72 to 6.31 ppm.Chloride levels were high in all samples ranging from 4000 to 4640 ppm (Table 2).

Sequence data
The raw data from the sediments and water samples ( three sediment and one water sample) consisted of 548,639 sequences, of which 334, 394 sequences were retained after removing sequences with different tags at each end for quality filtering and denoising.After removing singletons, chimeric sequences and OTUs of non-fungal organisms (<200 base pairs after phred20based quality trimming, sequences with ambiguous base calls, and those with homopolymer runs exceeding 6 bp), a total of 151 fungal OTUs recovered at 3% genetic distance, were included in the final analysis.

Composition and diversity of fungal communities
Based on BLASTn searches in SILVA SSU Reference 119 database, 151 OTUs were identified, most of which had their best matches against accessions in SILVA database.These 151 OTUs spanned 5 phyla namely; Ascomycota, Basidiomycota, Fungi unspecified phylum, Chytridiomycota and Glomeromycota.Sediment samples collected from 81°C had the highest number of OTUs (135 OTUs) while 88 OTUs were shared among all sample types (Figure 1).The shared OTUs were distributed among the phyla; Ascomycota (up to 42.9% relative abundance in sediment sample at 45.1°C), Fungi unspecified phylum (up to 6.2% relative abundance in sediment sample at 83.6°C), Basidiomycota (up to 3.3% relative abundance in sediment sample at 81°C), Chytridiomycota and Glomeromycota (up to 1.5% relative abundance each in water sample at 81°C.OTUs belonging to the Phylum Ascomycota were the most abundant and were represented    vegae284 each scoring a relative abundance of 6.25%.Antarctic fungal sp.gi944 of Fungi unspecified phylum also scored a relative abundance of 6.25% (Figure 2).The water sample collected at 81°C was found to harbor a higher diversity of fungi with low species richness as shown in Figure 2. Hierarchical clustering between samples collected from Lake Magadi revealed sediment samples from the two hot springs in Little Magadi "Nasikie eng'ida" to be closer than the sample from the hot spring in the main lake.Majority of the groups at species level included Aspergillus, Ascomycota, Penicillium, Neurospora, Termitomyces, Malassezia, Trichocomaceae, Stagonospora, Ramularia and Hypocreales (Figure 3).The dendogram shows relationship between the four samples.

Fungal richness and diversity indices
Richness (S) estimated the water sample (81°C) to be the richest site, constituting 35 taxa.Sediment samples from the three sites had Evenness (J') scores close to 1 (0.663 -0.897), hence showing evenness in their number of taxa members than the water sample (81°C).Simpson (1/D) also indicated the sediment 45.1°C to harbor the most diverse taxa (9.98).The Shannon's index (H' = 1.28-2.63)indicated low variation in the level of diversity among the sediment and water samples (Table 3).Analysis of similarity and distance based redundancy analysis at genus level showed connectivity of distance matrix with threshold dissimilarity of 1 indicating that data of the four samples are connected ([1] 1 1 1 1), hence there were no significant differences in community structure in the samples at 95% level of confidence (P value=0.05).

DISCUSSION
The significance of fungal communities in the hot springs of hypersaline lakes is unclear, mainly because data on the fungal species in these habitats is limited.Using traditional culture-based methods, researchers reported relatively low levels of diversity for fungal communities in extreme environments (Salano, 2011).In this study we used high-throughput sequencing technology in order to comprehensively analyze fungal communities within the hot springs.The high sensitivity of Illumina sequencing enables the detection of rare species, thus provides more detailed information on fungal diversity in these habitats.Members of Ascomycota were more frequently identified in the hot springs than those of Basidiomycota, whereas members of Chytridiomycota and Glomeromycota represented only a small proportion of the hot spring fungal communities.The abundance of Ascomycota is similar to the abundance of fungi determined in the previous study on soda soils at the edge of several lake basins, where filamentous fungi that could grow at high ambient pH values were isolated (Alexey et al., 2015).The results in that study revealed 100 strains of fungi with various degrees of alkali tolerance and taxonomic affinity within Ascomycota (Alexey et al., 2015).Additionally, 6.2% of the fungi detected in wet sediment 83.6°C were unspecified Phylum.These may be undiscovered and possibly indigenous species in the hot springs.Some of the groups in this study are similar to those recovered from a previous culture dependent study conducted on the Hot spring in main Lake Magadi (Salano, 2011) (Salano, 2011).Filamentous fungi like Aspergillus and Penicillium are attractive organisms for production of useful protein and biological active secondary metabolites.They have high secretion capacity and are effective hosts for the production of foreign proteins (Tsukagoshi et al., 2001).
Penicillium genera were found in wet sediments 45.1 and 81°C with relative abundance of 14.29 and 3.28% respectively.This is similar to previous studies in hypersaline water of salterns that revealed different species of Aspergillus, Penicillium and diverse nonmelanised yeasts (Gunde-Cimerman et al., 2005).Another study that used morphological and molecular techniques to identify a series of halotolerant fungi from hypersaline environments of solar salterns revealed 86 isolates of 26 species from salt ponds, which were identified as Cladosporium cladosporioides, nine Aspergillus sp., five Penicillium sp. and the black yeast Hortaea werneckii (Cantrell et al., 2006).Rhodotorula mucilaginosa, a yeast species and Rhizopus sp.30795, a Zygomycota were found in wet sediment at 81°C while unclassified Antarctic fungal sp.gi944 dominated wet sediment at 83.6°C.Other plant pathogenic fungi recovered included Fusarium sp., Cladosporium cladosporioides, Aspergillus flavus, Aspergillus japonicas and Aspergillus oryzae.Most of these organisms may have found their way to the hot springs through various dispersal mechanisms or may be adapted in these extreme environments.
According to Frontier (1985), harsh environments experiencing one or more extreme conditions tend to harbor fewer species.In contrast, wet sediments at hot spring 45.1°C were found to have the least OTUs (107 OTUs) as compared to higher temperature samples, distributed within Aspergillus oryzae (42.86%),Penicillium sp.5/97_5 (42.28%) and Trichocomaceae sp.lm65 (42.28%).Although water samples at 81°C were found to harbor a higher diversity of fungi with lower species richness, wet sediments showed a lower diversity with high abundance of present groups.This could be due to high abundance of organic matter and lower oxygen levels which favored decomposition processes; hence the groups present have sufficient carbon sources (Neira et al., 2001;Buee et al., 2009).The widespread fungal groups within the wet sediments may therefore be degraders of organic matter (Edgcomb et al., 2011a;Nagahama et al., 2011;Burgaud et al., 2013;Coolen et al., 2013).
This study reveals the presence of moderate and weak alkalitolerant fungi such as Alternaria alternata, Penicillium sp., Cladosporium sp. and Fusarium sp.previously reported to grow optimally at neutral or below neutral pH values.These species have previously appeared in existing reports on the alkalitolerant and halotolerant fungi (Kladwang et al., 2003;Gunde-Cimerman et al., 2009).They have therefore been considered as transition species in the alkaline environments, since they are also known to inhabit neutral soils worldwide.Hypocreales and Pleosporaceae have been reported as strong alkalitolerants and effective alkaliphiles inhabiting soda soils at the edge of lake basins (Alexey et al., 2015).In this study, Hypocreales sp.lm 566 was identified in water samples at 81°C while Pleosporales sp. was found in wet sediment and water samples at 83.6 and 81°C respectively.Other interesting groups recovered from this study include Pestalotiopsis sp., Neurospora sp., and Xylariaceae sp.These have been reported to have various applications in Biotechnology industries (Russell et al., 2011;Roche et al., 2014;Healy et al., 2004;Posada et al., 2007).

Conclusion
This study presented fungal diversity analysis of samples collected from the hot springs of Lake Magadi and Little Magadi, using Illumina Sequencing Technology.The results revealed representatives of thermophilic and alkaliphilic fungi within the hot springs, suggesting their ability to adapt to a multi-extreme sampling environment due to high pH, temperature, and salinity.Culture dependent studies in future will help us unravel the survival mechanisms used by these polyextremophilic fungi.

Figure 1 .
Figure 1.Venn diagram showing the distribution of unique and shared OTUs within various sample types in the three sampling sites.The number of OTUs in each hot spring is indicated in the respective circle.

Figure 3 .
Figure 3. Hierarchical clustering of DNA samples collected from the three hot springs of Lake Magadi and Little Magadi.Species level was chosen to be used in hierarchical clustering to assess the relationships between samples and taxa.

Table 1 .
Physico-chemical parameters of sampling stations in Lake Magadi and Little Magadi measured before sampling.

Table 2 .
Chemical analyses of samples from the Hot Springs Lake Magadi and Little Magadi.

Table 3 .
Diversity indices computed on all OTU-based fungal taxonomic units obtained from samples collected from the hot springs of Lake Magadi and Little Magadi.